Assessing the reliability of pre-sale price estimates: empirical evidence on Ferrari cars auctions

Details

Supervisors

Giot, Pierre

Faculty

Faculté des sciences économiques, sociales, politiques et de communication

Degree label

Master [120] en sciences économiques, orientation générale

Abstract

The pre-sale price estimates play an important role in the auction process. The experts that specialize in particular relevant field evaluate the goods before the actual sale will take place and propose the range of expected prices at which these goods might be sold, in form of minimum and maximum expected prices. The price estimates provided by auction houses are based on the information on characteristics of the good that is available to experts and should reflect its intrinsic value. Therefore, in efficient market the hammer prices should fall within the forecasted range. However, in reality the actual ex-post prices often do not coincide with those ex-ante estimated, they tend to either under- or overshoot the pre-estimated range. The main objective of this study is to analyze the deviations of hammer prices from forecasted ones and what possibly causes this phenomenon. By using the dataset that was constructed basing on the data from eight editions of “Classic Car Auction Yearbook” on the Ferrari cars auction sales we aim to verify the unbiasedness and reliability of pre-sale estimates provided by auction houses, assess the level of their uncertainty and accuracy in predictions and analyze by what factors these are affected. We find that pre-sale estimates provided by auction houses are biased, although extend of bias is rather small. Almost all major auction houses systematically undervalue cheap and overvalue expensive cars. This finding of our study significantly contrasts with the empirical findings of existing studies that in the majority of cases agree on the opposite: auction houses tend to overvalue inexpensive pieces and undervalue expensive ones or to generally undervalue their lots. With regard to vehicle characteristics, some of them have significant effect on final price, uncertainty and accuracy of prediction. We also find the evidence that supports the assumption that experts tend to set upper and lower bounds of the price range according to the rule of thumb and conventions, rather than being guided by strong economic reasoning. The unbiasedness and precision of pre-sales estimates constitute a relevant research question also because in the majority of cases they turn up to be the sole reference point in the valuation process for inexperienced buyers and might heavily affect the reserve price of a consignor.